Application of convolutional neural network-based endoscopic imaging in esophageal cancer or high-grade dysplasia: A systematic review and meta-analysis

BACKGROUND Esophageal cancer is the seventh-most common cancer type worldwide,accounting for 5%of death from malignancy.Development of novel diagnostic techniques has facilitated screening,early detection,and improved prognosis.Convolutional neural network(CNN)-based image analysis promises great potential for diagnosing and determining the prognosis of esophageal cancer,enabling even early detection of dysplasia.METHODS PubMed,EMBASE,Web of Science and Cochrane Library databases were searched for articles published up to November 30,2022.We evaluated the diagnostic accuracy of using the CNN model with still image-based analysis and with video-based analysis for esophageal cancer or HGD,as well as for the invasion depth of esophageal cancer.The pooled sensitivity,pooled specificity,positive likelihood ratio(PLR),negative likelihood ratio(NLR),diagnostic odds ratio(DOR)and area under the curve(AUC)were estimated,together with the 95%confidence intervals(CI).A bivariate method and hierarchical summary receiver operating characteristic method were used to calculate the diagnostic test accuracy of the CNN model.Meta-regression and subgroup analyses were used to identify sources of hetero-geneity.RESULTS A total of 28 studies were included in this systematic review and meta-analysis.Using still image-based analysis for the diagnosis of esophageal cancer or HGD provided a pooled sensitivity of 0.95(95%CI:0.92-0.97),pooled specificity of 0.92(0.89-0.94),PLR of 11.5(8.3-16.0),NLR of 0.06(0.04-0.09),DOR of 205(115-365),and AUC of 0.98(0.96-0.99).When video-based analysis was used,a pooled sensitivity of 0.85(0.77-0.91),pooled specificity of 0.73(0.59-0.83),PLR of 3.1(1.9-5.0),NLR of 0.20(0.12-0.34),DOR of 15(6-38)and AUC of 0.87(0.84-0.90)were found.Prediction of invasion depth resulted in a pooled sensitivity of 0.90(0.87-0.92),pooled specificity of 0.83(95%CI:0.76-0.88),PLR of 7.8(1.9-32.0),NLR of 0.10(0.41-0.25),DOR of 118(11-1305),and AUC of 0.95(0.92-0.96).CONCLUSION CNN-based image analysis in diagnosing esophageal cancer and HGD is an excellent diagnostic method with high sensitivity and specificity that merits further investigation in large,multicenter clinical trials.

Study on thermal-state variation of high-grade highway embankment under different pavement conditions

The research shows that the selection of pavement type is very important for the thermal stability of high-grade highway embankment in permafrost regions because of the different solar absorption rates between asphalt concrete and asphalt concrete pavement.In this paper,the common embankment of high-grade highway in permafrost regions is selected as the research object to study the influence of asphalt concrete and cement concrete pavement on the embankment temperature,freeze-thaw cycle process and the change law of the permafrost table,which provides a basis for the use of reasonable pavement materials in permafrost regions.

Lessons from construction and health condition evaluation of high-grade highway in permafrost regions

Located in the eastern margin of the Tibetan Plateau,the Gonghe-Yushu high-grade highway was the first of its kind in plateau permafrost regions.Most of the road sections along the high-grade highway are unstable or extremely unstable warm permafrost with an average annual ground temperature above −1℃,which is vulnerable to global warming and human engineering activities.This paper describes permafrost characteristics,roadbed design,and operation of the Gonghe-Yushu high-grade highway in detail.It is found that thaw settlement of warm and ice-rich permafrost is the main cause of subgrade subsidence in permafrost sections of this highway due to insufficient permafrost survey and drainage design.It is recommended that the interception and drainage system's design be optimized,and the permafrost upper limit and the variation of ground temperature be further investigated to provide essential data for the treatment of highway distress.It should be emphasized that protecting permafrost soil environment and optimized engineering design are crucial to successful high-grade highway engineering in permafrost regions.

Identification of critical links under earthquake hazards for highway networks

Critical links are defined as easily damaged links with massive transport in highway networks, which also need intensive improvement. The total travel time increment caused by a link＇s failure reflects its importance and is taken as the measure of importance. Links are subdivided into segments according to their structure features and environments. Each segment＇s unreliability is the probability of its function failure that cannot be recovered within an expected time. The measure of criticality is defined as the expected total travel time increment and can be obtained from the product of importance and reliability. It reflects a links＇ importance and ability to provide continuous service for evacuation and rescues under earthquake situation. Critical links can then be identified from the sequence of their criticality. These measures are calculated in the highway network of earthquake-hit areas in Wenchuan. Results collected in geographic information system （GIS） visualization are consistent with the situation revealed in this earthquake, which indicates that the presented method can be used to identify critical links in advance and give guidance regarding refugee evacuation and facility protection from earthquakes.

The Application of Trapezoidal Fuzzy AHP on Evaluation Item Weight Calculation of Highway Network Programming Schemes

The fuzzy theory is invited in this paper to calculate weights of evaluation items, which is a very important stage in highway network evaluation. Highway network evaluation is a composite of various attributes that lack of pertinency and comparability, and their commonality is that they are more or less tainted with uncertainty, with different daily decision making problems of diverse intensity, the results can be misleading if the fuzziness of human decision-making is not taken into account. So we introduce fuzzy theory into the measurement of evaluating weights to overcome the issue and Analytic Hierarchy Process in scheme ranking. It is testified to be a more reliable way through the case study of Yangtze River Delta highway network evaluation.

Geohazards and risk assessment along highway in Sichuan Province, China

Geohazards along highways are the main natural hazards that could affect the safety and operation of highway systems.Understanding the risks faced by highways in areas affected by geohazards is an urgent problem to be solved.This study used historical geohazard events from Sichuan Province and highway network data to propose a geohazard risk index that reflects the risk geohazards along highways.Furthermore,this work applied the entropy method and expert scoring to calculate the weight of the index.The spatial distributions of landslides,debris flows,collapses,and unstable slopes along the highways were analysed based on ArcGIS spatial statistics,and the highway geohazard intensity index were obtained.The relationships between slope,rainfall,vegetation coverage,rock type,land use,and incision depth with geohazards were analysed,and the highway geohazard susceptibility index was calculated by the weighted information method.Based on the intensity and susceptibility index,we obtained a geohazard risk index which can better evaluate the risk of highways,and made a highway geohazard risk map to aid the prevention and mitigation of geohazards along highways and assist with highway network planning.

Modeling Highway Traffic Safety for Developing Countries Using Delphi Technique and Bayesian Network： Case Study for Nigeria

Highway traffic safety is an issue confronting developing countries and those of industrialized nations. Nigeria, as a developing country, has been experiencing unusually high traffic related injury and fatality rates. In order to commence combating highway safety problems in Nigeria, the first task is to identify the major contributing factors; however, Nigeria has no reliable and comprehensive database of traffic accidents and casualties. Consequently, the Delphi technique was utilized in generating the required data such as number of registered automobiles, number of licensed drivers, and annual fatality count for modeling and forecasting accident rates in Nigeria. A Bayesian network model was developed and used, with the data obtained from Delphi process, to demonstrate possible traffic safety responses to different scenarios of changes in the Nigerian socio-political culture. Although the Delphi technique and the Bayesian network model only estimate the accident and safety data, those methods can be a realistic option when those data are not available, especially for the developing countries. As a result, the major accident contributors have been identified and the top three contributors-road condition, DUI （driving under the influence） and reckless driving-are policy related. The Nigerian traffic safety outlook would improve significantly if the existing laws and policies can be enforced, even at a very moderate level.

The Analysis of Spatial-Temporal Evolution of City Accessibility Based on Highway Network in Henan Province in China

Accessibility is an important tool to evaluate the maturity of a regional traffic network structure which describes the traffic convenience in the traffic network. The paper defines a new accessibility index by using the resident population weighted average value of the sum of inverse of the traveling time distance and time threshold coming from ordinary traffic network, and then uses this accessibility index to analyze the spatial-temporal characteristics of Henan highway network, as well as its evolution patterns from 2005 to 2020. The results show that with the expansion and improvement of Henan highway network, city accessibility level has been significantly improved, spatial convergence is obvious, the cities in the north central are always High-High aggregation area, the cities in the south are always Low-Low aggregation area, gradually forming the characteristics of Northwest high and Southeast low, relative balance between East and West. There is some non-conforming phenomenon in highway mileage growth and improvement of the city accessibility levels, but this situation is being weakened, the highway network layout is gradually rationalized, the spatial distribution of city accessibility and that of population are beginning to converge.

A Novel Network Screening Methodology for Rural Low-Volume Roads

Low-volume roads (LVRs) are an integral part of the rural transportation network providing access to remote rural areas and facilitating the movement of goods from farms to markets. These roads pose unique challenges for highway agencies including those related to safety management on the highway network. Specifically, traditional network screening methods using crash history can be effective in screening rural highways with higher traffic volumes and more frequent crashes. However, these traditional methods are often ineffective in screening LVR networks due to low traffic volumes and the sporadic nature of crash occurrence. Further, many of the LVRs are owned and operated by local agencies that may lack access to detailed crash, traffic and roadway data and the technical expertise within their staff. Therefore, there is a need for more efficient and practical network screening approaches to facilitate safety management programs on these roads. This study proposes one such approach which utilizes a heuristic scoring scheme in assessing the level of risk/safety for the purpose of network screening. The proposed scheme is developed based on the principles of US Highway Safety Manual (HSM) analysis procedures for rural highways and the fundamentals in safety science. The primary application of the proposed scheme is for ranking sites in network screening applications or for comparing multiple improvement alternatives at a specific site. The proposed approach does not require access to detailed databases, technical expertise, or exact information, making it an invaluable tool for small agencies and local governments (e.g. counties, townships, tribal governments, etc.).

Small object detection on highways via balance feature fusion and task-specific encoding network

Detecting small objects on highways is a novel research topic.Due to the small pixel of objects on highways,traditional detectors have difficulty in capturing discriminative features.Additionally,the imbalance of feature fusion methods and the inconsistency between classification and regression tasks lead to poor detection performance on highways.In this paper,we propose a balance feature fusion and task-specific encoding network to address these issues.Specifically,we design a balance feature pyramid network(FPN)to integrate the importance of each layer of feature maps and construct long-range dependencies among them,thereby making the features more discriminative.In addition,we present task-specific decoupled head,which utilizes task-specific encoding to moderate the imbalance between the classification and regression tasks.As demonstrated by extensive experiments and visualizations,our method obtains outstanding detection performance on small object detection on highways(HSOD)dataset and AI-TOD dataset.

Monitoring subsidence rates along road network by persistent scatterer SAR interferometry with high-resolution TerraSAR-X imagery

Ground subsidence is one of the key factors damaging transportation facilities, e.g., road networks consisting of highways and railways. In this paper, we propose to apply the persistent scatterer synthetic aperture radar interferometry （PS-InSAR） approach that uses high- resolution TerraSAR-X （TSX） imagery to extract the regional scale subsidence rates （i.e., average annual sub- sidence in mm/year） along road networks. The primary procedures involve interferometric pair selection, interfer- ogram generation, persistent scatterer （PS） detection, PS networking, phase parameterization, and subsidence rate estimation. The Xiqing District in southwest Tianjin （China） is selected as the study area. This district contains one railway line and several highway lines. A total of 15 TSX images covering this area between April 2009 and June 2010 are utilized to obtain the subsidence rates by using the PS-InSAR （PSI） approach. The subsidence rates derived from PSI range from -68.7 to -1.3 mm/year. These findings show a significantly uneven subsidence pattern along the road network. Comparison between the PSI-derived subsidence rates and the leveling data obtained along the highways shows that the mean and standard deviation （SD） of the discrepancies between the two types of subsidence rates are 0.1 and 4-3.2 mm/year, respectively. The results indicate that the high-resolution TSX PSI is capable of providing comprehensive and detailed subsidence information regarding road networks with millimeter-level accuracy. Further inspections under geo- logical conditions and land-use categories in the study area indicate that the observed subsidence is highly related to aquifer compression due to groundwater pumping. Therefore, measures should be taken to mitigate groundwater extraction for the study area.

Application of artificial neural networks for operating speed prediction at horizontal curves: a case study in Egypt

Horizontal alignment greatly affects the speedof vehicles at rural roads. Therefore, it is necessary toanalyze and predict vehicles speed on curve sections.Numerous studies took rural two-lane as research subjectsand provided models for predicting operating speeds.However, less attention has been paid to multi-lane highwaysespecially in Egypt. In this research, field operatingspeed data of both cars and trucks on 78 curve sections offour multi-lane highways is collected. With the data, correlationbetween operating speed （V85） and alignment isanalyzed. The paper includes two separate relevant analyses.The first analysis uses the regression models toinvestigate the relationships between V85 as dependentvariable, and horizontal alignment and roadway factors asindependent variables. This analysis proposes two predictingmodels for cars and trucks. The second analysisuses the artificial neural networks （ANNs） to explore theprevious relationships. It is found that the ANN modelinggives the best prediction model. The most influential variableon V85 for cars is the radius of curve. Also, for V85 fortrucks, the most influential variable is the median width.Finally, the derived models have statistics within theacceptable regions and they are conceptually reasonable.

Optimal Retrofit Scheme for Highway Network under Seismic Hazards

Many older highway bridges in the United States(US)are inadequate for seismic loads and could be severely damaged or collapsed in a relatively small earthquake.According to the most recentAmerican Society of Civil Engineers’infrastructure report card,one-third of the bridges in the US are rated as structurally deficient and many of these structurally deficient bridges are located in seismic zones.To improve this situation,at-risk bridges must be identified and evaluated and effective retrofitting programs should be in place to reduce their seismic vulnerabilities.In this study,a new retrofit strategy decision scheme for highway bridges under seismic hazards is developed and seamlessly integrate the scenario-based seismic analysis of bridges and the traffic network into the proposed optimization modeling framework.A full spectrum of bridge retrofit strategies is considered based on explicit structural assessment for each seismic damage state.As an empirical case study,the proposed retrofit strategy decision scheme is utilized to evaluate the bridge network in one of the active seismic zones in the US,Charleston,South Carolina.The developed modeling framework,on average,will help increase network throughput traffic capacity by 45%with a cost increase of only$15million for the Mw 5.5 event and increase the capacity fourfold with a cost of only$32m for the Mw 7.0 event.

Accessibility for High Asia:Comparative Perspective on Northern Pakistan's Traffic Infrastructure and Linkages with Its Neighbours in the Hindukush- Karakoram-Himalaya

Starting with a discussion of development concepts which were applied in practice and which followed the developmentalist paradigm the expansion of traffic infrastructure in colonial and post-colonial periods is presented for the High Asian mountain rim. Selective railways and roads are the major feature of this development, which aimed first on serving the convenience of hill station visitors and followed strategic considerations later on. This bias between regional planning and implementation remains a characteristic feature. At the same time traffic infrastructure without asphalt roads is important for the mountain areas, thus breaking up the strong correlation between development and asphalt roads.

Assessment of highway slope failure using neural networks

An artificial intelligence technique of back-propagation neural networks is used to assess the slope failure.On-site slope failure data from the South Cross-Island Highway in southern Taiwan are used to test the performance of the neural network model.The numerical results demonstrate the effectiveness of artificial neural networks in the evaluation of slope failure potential based on five major factors,such as the slope gradient angle,the slope height,the cumulative precipitation,daily rainfall and strength of materials.

Economic Assessment of Planned Projects in Highway Network Planning

This paper presents a method of economic assessment for planned projects in the process of highway network planning. Economic assessment method is being done on the basis of cost benefit analysis, and the cost and benefit calculation methods are discussed respectively.

For Whom the Toll Sells

For decades,China's extensive network of highways has largely remained toll-free However,local governments are pushing harder than ever before for tolls to heln cover costs and pull them out of debt.

Dysfunction of PLA2G6 and CYP2C44-associated network signals imminent carcinogenesis from chronic inflammation to hepatocellular carcinoma

Little is known about how chronic inflammation contributes to the progression of hepatoceUular carcinoma （HCC）, especially the initiation of cancer. To uncover the critical transition from chronic inflammation to HCC and the molecular mechanisms at a network level, we analyzed the time-series proteomic data of woodchuck hepatitis virus/c.myc mice and age-matched wt-C57BL/6 mice using our dynamical network biomarker （DNB） model. DNB analysis indicated that the 5th month after birth of transgenic mice was the critical period of cancer initiation, just before the critical transition, which is consistent with clinical symptoms. Meanwhile, the DNB-associated network showed a drastic inversion of protein expression and coexpression levels before and after the critical transition. Two members of DNB, PLA2G6 and CYP2C44, along with their associated differentially expressed proteins, were found to induce dysfunction of arachidonic acid metabolism, further activate inflammatory responses through inflammatory mediator regulation of transient receptor potential channels, and finally lead to impairments of liver detoxification and malignant transition to cancer. As a c-Myc target, PLA2G6 positively correlated with c-Myc in expression, showing a trend from decreasing to increasing during carcinogenesis, with the minimal point at the critical transition or tipping point. Such trend of homologous PLA2G6 and c-Myc was also observed during human hepatocarcinogenesis, with the minimal point at high-grade dysplastic nodules （a stage just before the carcinogenesis）. Our study implies that PLA2G6 might function as an oncogene like famous c-Myc during hepatocar- cinogenesis, while downregulation of PLA2G6 and c-Myc could be a warning signal indicating imminent carcinogenesis.

Study on the influence of skid resistance on traffic safety of highway with a high ratio of bridges and tunnels

The safety of highways with a high ratio of bridges and tunnels is related to multiple factors,for example,the skid resistance of the pavement surface.In this study,the distribution of accidents under different conditions was calculated to investigate the relationship between the road skid resistance and the incidence of traffic accidents based on the traffic accident data of the Yuxiang highway.Statistical results show that weather conditions and road alignment may affect traffic accidents.The correlation analysis method was used to study the relationship between three factors and traffic accidents.The results show that road alignment,weather conditions and road skid resistance are related to the incidence of traffic accidents.The traffic accident prediction models were established based on back propagation neural network to verify the correlation analysis results.It is confirmed that road alignment,weather conditions and road skid resistance are the factors that affect traffic accidents.

Towards better entity linking

As one of the most important components in knowledge graph construction,entity linking has been drawing more and more attention in the last decade.In this paper,we propose two improvements towards better entity linking.On one hand,we propose a simple but effective coarse-to-fine unsupervised knowledge base(KB)extraction approach to improve the quality of KB,through which we can conduct entity linking more efficiently.On the other hand,we propose a highway network framework to bridge key words and sequential information captured with a self-attention mechanism to better represent both local and global information.Detailed experimentation on six public entity linking datasets verifies the great effectiveness of both our approaches.